Rain/thunder pushpin contrary to table and accumulation

[mdteason]

This might be insignificant, or even intentional, given the many challenges parsing weather models at various timeframes, but today, the (3h) Rain/thunder accumulation forecasts reported at the map pushpin was quite inconsistent (about double) with the table reached via the dropdown arrow. A specific example from now (15:00 PST, December 10, 2022) is that, scrolling forward to the 0600 PST, Dec. 11 timeframe, the pushpin forecasts 0.47 inch rain accumulation. The description of weather overlays says that 0.47 inch is forecast to have occurred in the prior three hours (from 0300 to 0600). The inconsistency is that the dropdown table forecasts only 0.22 inch for that timeframe. Is it possible that the pushpin might be accidentally forecasting a timeframe longer than 3 hours, or different start/stop times?
The table is consistent with the other, forward-looking 24-hours rain accumulation, but it is also possible that those indicators are somehow not accumulating as intended.
If I am just confused, as usual, sorry! If I am right, please consider it minor constructive criticism. Windy is very useful and fun. Thank you. Mike Dean, California, USA.

[idefix37]

@mdteason
I think the explanation for this discrepancy is because the Rain accumulation displayed by the weather picker is computed from the principal reference times (run time) of the model i.e. 00h UTC / 12h UTC.
Here is the Basic meteogram shown in California with the 00h UTC update of the model:

In this diagram, hours are not shown from the reference time of the model. If rain is predicted during these missing hours, the Rain accumulation will be higher.
And in California the gap between UTC and local time is important…